July 17 news.OpenAI Announced via the X account, the live stream will take place on the 17th at 10 a.m. PT (note: 1 a.m. BST on the 18th).
The official teaser image released shows a mouse pointer showing a pentagonal trajectory, causing netizens to wonderLatest Models GPT-5 expectations.
As 1AI previously reported, OpenAI CEO Ortmann personally revealed the timeline for the release of GPT-5 on a podcast, further reiterating that the "most powerful model" will be released this summer. GPT-5 is not just about performance, it's about being an "all-in-one" model. Users will no longer need to switch between GPT-4 for text processing and DALL-E for image generation; GPT-5 will combineText, image, voice and possibly even video processing capabilitiesAll in one, providing a unified interface for use.
OpenAI says that the goal is toSimplifying the user experience while providing consistency. Romain Huet, who is responsible for developer experience, said that GPT-5 willIntegrate the strengths of past generations of models to become more powerful and easier to use at the same time.
GPT-5 is expected to support larger context windows, allowing for longer conversations and more complex tasks. In addition, it will be able to "self-adjust" based on user behavior, so that over time the answers will be more responsive to individual needs.